In Reply

We are grateful to Drs. Priebe, Xue, and Weiskopf for their interest regarding our manuscript entitled Lack of Nephrotoxicity by 6% Hydroxyethyl Starch 130/0.4 during Hip Arthroplasty: A Randomized Controlled Trial,¹  which appeared in the November 2014 issue of Anesthesiology. Further, we thank for their complimentary words and remarks. We will answer the queries starting with Dr. Priebe, then Dr. Xue and finally Dr. Weiskopf.

Dear Dr. Priebe, you request additional information regarding urine 4 in terms of plasma creatinine and creatinine clearance. However, urine 4 was obtained just before discharge and was a “spot urine,” i.e., not a urine collection over time. Thus, creatinine clearance could not be calculated. Additional analyses of urine and blood samples could have contributed with further information in the postoperative period and during follow-up, but this was not included in the protocol.

The study design allowed us to compare the effect 6% hydroxyethyl starch (HES) 130/0.4 and isotonic saline 0.9% on urinary neutrophil gelatinase-associated lipocalin (u-NGAL) during hip arthroplasty. We can conclude that no difference existed between the two solutions, but an increase was seen in u-NGAL in both infusion groups in urine 4. We used isotonic saline 0.9% as control fluid, because it had the same chloride content of 154 mmol similar to the intervention fluid. We agree that a possible nephrotoxic effect of the chloride component in isotonic saline 0.9% is interesting, and further studies are necessary to clarify this aspect. However, other studies that were comparing a balanced solution, i.e., lactated Ringer’s solution or similar to a chloride-rich solution, found no differences in u-NGAL in the groups.^2,3 

Dear Dr. Xue, we used a cutoff value of 100 ng/ml for u-NGAL. We wanted to see whether 6% HES 130/0.4 inflicted none, mild, or severe renal injury compared with isotonic saline 0.9%. There were nine versus seven patients in the HES versus saline group with a u-NGAL value more than 100 ng/ml at discharge. Thus, no difference existed between the groups.

The study was not designed to compare the occurrence of acute kidney injury (AKI) between HES 6% 130/0.4 and isotonic saline 0.9%. It goes without saying that a huge number of subjects had to be included, if AKI should be the primary effect variable. We agree that fulminant AKI is a seldom event after noncardiac surgery, but the outcome in our study was differences in renal markers specific for renal injury, i.e., u-NGAL, plasma creatinine, urine output, and creatinine clearance. So, our study was powered to find a difference in these markers and not to find a difference in the incidence of AKI. When evaluating HES-induced renal failure, it is important to differentiate between a surgical population and a septic one. The findings in severe sepsis are not applicable to surgical patients.⁴  Further, there are numerous pharmacokinetic differences between the generations of starches and the findings of side effects. The elder generations of starches cannot be compared with the latest generations of starch.⁵  Until now, no evidence exists of a perioperative renal impairment after tetrastarch infusion in subjects with normal renal function before surgery.^6–8 

In a previous study, the follow-up was 28 days after HES infusion, and no signs were detected of HES-induced renal impairment.⁹  We are convinced that we would have seen signs suggestive of renal injury within the 14 days of follow-up in the present study, if there had been any.

Dear Dr. Weiskopf, thank you for the complementary words. Due to space limitations, we did not publish the absolute or relative changes from baseline to the follow-up. Among the absolute changes between baseline and follow-up, we found only a significant difference between the groups for creatinine clearance (P < 0.01), whereas u-NGAL, u-NGAL_CR (urinary neutrophil gelatinase-associated lipocalin adjusted for creatinine), and urine output did not deviate significantly. Among the relative changes, calculated as (follow-up − baseline)/follow-up × 100, only creatinine clearance deviated significantly (P = 0.02), whereas no significant differences were found in u-NGAL, u-NGAL_CR, and urine output. Thus, creatinine clearance increased slightly in the HES group and was unchanged in the saline group.